| 1 | Flux.1-dev-Controlnet-Upscalerjasperai | Image To Image | 2 | ▲0.1%(up) | 888 | 1.8B |
| 2 | Qwen3-TTS-GGUFServeurperso | Text to Speech | 2 | ▲4.9%(up) | 63 | — |
| 3 | MiniMax-H3-RemixFX-FeiHou | Image To Video | 2 | — | 32 | — |
| 4 | whisper-small.enopenai | Speech Recognition | 2 | ▲0.8%(up) | 78 | 242M |
| 5 | NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-MLX-4bitrapid-mlx | Text Generation | 2 | — | 2 | 121B |
| 6 | swin-small-patch4-window7-224microsoft | Image Classification | 2 | ▲0.3%(up) | 8 | — |
| 7 | Yue2-3B-GGUFaudio-cpp | Text to Audio | 2 | ▲253.5%(up) | 99 | — |
| 8 | Magma-8Bmicrosoft | Robotics | 2 | ▲0.3%(up) | 424 | 8.9B |
| 9 | xclip-base-patch16-hmdb-2-shotmicrosoft | Video Classification | 2 | — | 5 | — |
| 10 | pplx-embed-v1-0.6bperplexity-ai | Embeddings | 2 | ▲0.3%(up) | 237 | 596M |
| 11 | GEM-Xnvidia | — | 2 | ▲3.3%(up) | 31 | — |
| 12 | Qwen2.5-Math-PRM-72BQwen | Text Classification | 2 | ▲0.2%(up) | 83 | 72.8B |
| 13 | DeepSeek-Coder-V2-Instructdeepseek-ai | Text Generation | 2 | ▲0.2%(up) | 718 | 236B |
| 14 | beit-base-patch16-384microsoft | Image Classification | 2 | ▲0.1%(up) | 10 | — |
| 15 | mask2former-swin-small-cityscapes-semanticfacebook | Image Segmentation | 2 | ▲0.4%(up) | 7 | 68.8M |
| 16 | DialoGPT-mediummicrosoft | Text Generation | 2 | ▲0.3%(up) | 449 | — |
| 17 | Qwen_Qwen3.5-4B-GGUFbartowski | Image Text to Text | 2 | ▲2.7%(up) | 71 | — |
| 18 | granite-vision-4.1-4bibm-granite | Image Text to Text | 2 | ▲10.1%(up) | 112 | 4B |
| 19 | MOSS-Audio-4B-InstructOpenMOSS-Team | Audio Text To Text | 2 | ▲1.4%(up) | 89 | 5.2B |
| 20 | Swift-1.5-Qwen3.8-Flash-Next-W4A16-AWQukisai | Image Text to Text | 2 | — | 4 | 180B |
| 21 | chinese-decision-0.8b-ggufquincyqiang | Zero-Shot Classification | 2 | — | 4 | — |
| 22 | Phoenix-X-26B-A4B-Heretic-GGUFmradermacher | — | 2 | 0% | 2 | — |
| 23 | trocr-base-handwrittenmicrosoft | Image to Text | 2 | ▲0.1%(up) | 532 | 333M |
| 24 | mms-tts-fonfacebook | Text to Speech | 2 | — | 10 | 36.3M |
| 25 | imagegpt-largeopenai | — | 2 | — | 54 | — |
| 26 | stable-cascadestabilityai | Text to Image | 2 | ▲0.2%(up) | 1,346 | — |
| 27 | Qwen3.8-cyber-RedTeam-Surgical-Abliteratedmedismera | Text Generation | 2 | — | 7 | 27.4B |
| 28 | LFM2.5-24B-A2B-APEX-I-NanoPlus-GGUFIsValorum | Text Generation | 2 | — | 4 | — |
| 29 | OpenELMapple | — | 2 | — | 1,460 | — |
| 30 | Qwen3.8-Flash-Next-Heretic2-IQ4XS-NGQ4-GGUFcygnal | Image Text to Text | 2 | — | 8 | — |
| 31 | Linear-50M-expUser01110 | — | 2 | — | 5 | 49.6M |
| 32 | Qwen3.8-Flash-Next-Sushi-3bpwbeamster | Image Text to Text | 2 | — | 11 | 28.8B |
| 33 | mms-tts-bqifacebook | Text to Speech | 2 | — | 2 | 36.3M |
| 34 | Qwen1.5-0.5B-Chat-GPTQ-Int8Qwen | Text Generation | 2 | — | 11 | 464M |
| 35 | Phi-3-medium-4k-instruct-onnx-cudamicrosoft | Text Generation | 2 | — | 15 | — |
| 36 | Mistral-Nemo-Instruct-FP8-2407mistralai | — | 2 | — | 34 | — |
| 37 | geneformer_V2_104Mnvidia | Fill Mask | 2 | — | 5 | 120M |
| 38 | Qwen3-4B-Instruct-2507-GGUFlmstudio-community | Text Generation | 2 | ▲5.5%(up) | 20 | — |
| 39 | JanusFlow-1.3Bdeepseek-ai | Any to Any | 2 | ▲0.1%(up) | 168 | 2B |
| 40 | Qwen2.5-1.5B-Instruct-uncensored-GGUFmradermacher | — | 2 | ▲3.9%(up) | 24 | — |
| 41 | pix2struct-basegoogle | Image to Text | 2 | ▲1.3%(up) | 90 | 282M |
| 42 | Dayhoff-170M-UR90-HL-44000microsoft | Text Generation | 2 | — | 4 | 170M |
| 43 | tmax-9ballenai | — | 2 | ▲1.8%(up) | 19 | 9B |
| 44 | qwen3_5_35b_a3b_base_ohsdk_200k_rlLego-X | Text Generation | 2 | — | 2 | 36B |
| 45 | YOLO-head-detectionwesjos | Object Detection | 2 | — | 3 | — |
| 46 | Ling-3.0-Tiny-DSpark-GGUFStefanoBalocco | — | 2 | — | 6 | — |
| 47 | metricx-24-hybrid-xxl-v2p6google | — | 2 | ▲2.8%(up) | 21 | — |
| 48 | Translation-EnKo_-_gemma-2-2b-it-translation-general1.2m-en-ko-e1-b16-trc313eval45-e1-b16-ggufRichardErkhov | — | 2 | — | 2 | — |
| 49 | t5-efficient-small-dm256google | — | 2 | — | 5 | — |
| 50 | t5-small-ssmgoogle | — | 2 | — | 4 | — |